Effect Size

An effect size answers the question a p-value cannot: how big is the difference or relationship? Because p depends heavily on sample size, a trivial effect turns significant in a large sample while a substantial one can miss significance in a small one. Journals increasingly require an effect size alongside every test.

The measure follows the test: Cohen's d for two-group mean differences, eta squared or partial eta squared for ANOVA, Pearson's r for correlation, R squared for regression, and odds ratios for logistic models. For d, 0.2, 0.5 and 0.8 are the conventional small, medium and large landmarks.

Treat those labels as rough guides rather than verdicts — what counts as meaningful depends on the discipline and the outcome being measured. Report the effect size with its confidence interval and interpret it in the units and context of your own study, not against a generic benchmark.